mlsanityData to 7 Oct 2026

Velox: Learning Representations of 4D Geometry and Appearance

arXiv 2605.04527 ↗First posted 6 May 2026cs.CVcited by 0

Not on the map. Why?

Abstract (arXiv)

We introduce a framework for learning latent representations of 4D objects which are descriptive, faithfully capturing object geometry and appearance; compressive, aiding in downstream efficiency; and accessible, requiring minimal input, i.e., an unstructured dynamic point cloud, to construct. Specifically, Velox trains an encoder to compress spatiotemporal color point clouds into a set of dynamic shape tokens. These tokens are supervised using two complementary decoders: a 4D surface decoder, which models the time-varying surface distribution capturing the geometry; and a Gaussian decoder, which maps the tokens to 3D Gaussians, helping learn appearance. To demonstrate the utility of our representation, we evaluate it across three downstream tasks -- video-to-4D generation, 3D tracking, and cloth simulation via image-to-4D generation -- and observe strong performances in all settings.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 103 of its 124 references that are in the map. First posted on arXiv 6 May 2026.

Builds on

103 of 124 references.

References · 103

  1. 2017Attention Is All You Needcited by 59,932
  2. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,456
  3. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,694
  4. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,267
  5. 2020A Simple Framework for Contrastive Learning of Visual Representationscited by 11,979
  6. 2018The Unreasonable Effectiveness of Deep Features as a Perceptual Metriccited by 11,029
  7. 2023DINOv2: Learning Robust Visual Features without Supervisioncited by 7,764
  8. 2019Momentum Contrast for Unsupervised Visual Representation Learningcited by 7,482
  9. 2021Emerging Properties in Self-Supervised Vision Transformerscited by 6,893
  10. 20233D Gaussian Splatting for Real-Time Radiance Field Renderingcited by 6,141
  11. 2022Scalable Diffusion Models with Transformerscited by 5,931
  12. 2020Bootstrap your own latent: A new approach to self-supervised Learningcited by 4,712
38 more
  1. 2017PointNet++: Deep Hierarchical Feature Learning on Point Sets in a Metric Spacecited by 4,384
  2. 2022FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awarenesscited by 3,611
  3. 2017Quo Vadis, Action Recognition? A New Model and the Kinetics Datasetcited by 3,259
  4. 2020Unsupervised Learning of Visual Features by Contrasting Cluster Assignmentscited by 2,845
  5. 2021Barlow Twins: Self-Supervised Learning via Redundancy Reductioncited by 1,818
  6. 2020GLU Variants Improve Transformercited by 1,724
  7. 2023DUSt3R: Geometric 3D Vision Made Easycited by 1,526
  8. 2022Objaverse: A Universe of Annotated 3D Objectscited by 1,522
  9. 2023Zero-1-to-3: Zero-shot One Image to 3D Objectcited by 1,483
  10. 2020D-NeRF: Neural Radiance Fields for Dynamic Scenescited by 1,459
  11. 2019Root Mean Square Layer Normalizationcited by 1,425
  12. 2018Towards Accurate Generative Models of Video: A New Metric & Challengescited by 1,204
  13. 20234D Gaussian Splatting for Real-Time Dynamic Scene Renderingcited by 1,144
  14. 2023MVDream: Multi-view Diffusion for 3D Generationcited by 928
  15. 2023Objaverse-XL: A Universe of 10M+ 3D Objectscited by 767
  16. 2023Deformable 3D Gaussians for High-Fidelity Monocular Dynamic Scene Reconstructioncited by 753
  17. 2023K-Planes: Explicit Radiance Fields in Space, Time, and Appearancecited by 742
  18. 2024Structured 3D Latents for Scalable and Versatile 3D Generationcited by 742
  19. 2021Diffusion Probabilistic Models for 3D Point Cloud Generationcited by 734
  20. 2023Dynamic 3D Gaussians: Tracking by Persistent Dynamic View Synthesiscited by 731
  21. 2021Perceiver IO: A General Architecture for Structured Inputs & Outputscited by 616
  22. 2022LION: Latent Point Diffusion Models for 3D Shape Generationcited by 552
  23. 2025Continuous 3D Perception Model with Persistent Statecited by 527
  24. 2025Mean Flows for One-step Generative Modelingcited by 524
  25. 2023CoTracker: It is Better to Track Togethercited by 516
  26. 2023Zero123++: a Single Image to Consistent Multi-view Diffusion Base Modelcited by 516
  27. 2019PointFlow: 3D Point Cloud Generation with Continuous Normalizing Flowscited by 487
  28. 20233DShape2VecSet: A 3D Shape Representation for Neural Fields and Generative Diffusion Modelscited by 449
  29. 2024CLAY: A Controllable Large-scale Generative Model for Creating High-quality 3D Assetscited by 444
  30. 2020Modular Primitives for High-Performance Differentiable Renderingcited by 439
  31. 2022Point Transformer V2: Grouped Vector Attention and Partition-based Poolingcited by 384
  32. 2022Fast Dynamic Radiance Fields with Time-Aware Neural Voxelscited by 344
  33. 2023ImageDream: Image-Prompt Multi-view Diffusion for 3D Generationcited by 256
  34. 2023Michelangelo: Conditional 3D Shape Generation based on Shape-Image-Text Aligned Latent Representationcited by 225
  35. 2018Learning Particle Dynamics for Manipulating Rigid Bodies, Deformable Objects, and Fluidscited by 219
  36. 2022Tensor4D : Efficient Neural 4D Decomposition for High-fidelity Dynamic Reconstruction and Renderingcited by 217
  37. 2022Particle Video Revisited: Tracking Through Occlusions Using Point Trajectoriescited by 215
  38. 2024Direct3D: Scalable Image-to-3D Generation via 3D Latent Diffusion Transformercited by 210

50 most cited of 103 All 103 (JSON)

Built on by

Not cited yet.